whisper-base-finetuned-300

This model is a fine-tuned version of openai/whisper-base on the audiofolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.0044
• Wer Ortho: 67.5676
• Wer: 67.5676

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 30
• training_steps: 300

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.0021	20.0	60	0.9642	67.5676	67.5676
0.0003	40.0	120	0.9834	70.2703	70.2703
0.0002	60.0	180	0.9981	67.5676	67.5676
0.0001	80.0	240	1.0037	67.5676	67.5676
0.0001	100.0	300	1.0044	67.5676	67.5676

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1